Accurate Inference for Adaptive Linear Models
arXiv:1712.06695
Abstract
Estimators computed from adaptively collected data do not behave like their non-adaptive brethren. Rather, the sequential dependence of the collection policy can lead to severe distributional biases that persist even in the infinite data limit. We develop a general method -- -decorrelation -- for transforming the bias of adaptive linear regression estimators into variance. The method uses only coarse-grained information about the data collection policy and does not need access to propensity scores or exact knowledge of the policy. We bound the finite-sample bias and variance of the -estimator and develop asymptotically correct confidence intervals based on a novel martingale central limit theorem. We then demonstrate the empirical benefits of the generic -decorrelation procedure in two different adaptive data settings: the multi-armed bandit and the autoregressive time series.
arXiv commentTypos fixed for clarification
References in corpus (8)
- A Contextual-Bandit Approach to Personalized News Article Recommendation
- Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
- Unbiased Offline Evaluation of Contextual-bandit-based News Article Recommendation Algorithms
- The KL-UCB Algorithm for Bounded Stochastic Bandits and Beyond
- Doubly Robust Policy Evaluation and Learning
- lil' UCB : An Optimal Exploration Algorithm for Multi-Armed Bandits
- Online Least Squares Estimation with Self-Normalized Processes: An Application to Bandit Problems
- Why Adaptively Collected Data Have Negative Bias and How to Correct for It